Applications of machine learning tools in radiotherapy

• Grant number: 21/02254-6
• Support Opportunities: Regular Research Grants
• Start date: August 01, 2021
• End date: July 31, 2023
• Field of knowledge: Physical Sciences and Mathematics - Physics - Nuclear Physics
• Principal Investigator: Juliana Fernandes Pavoni
• Grantee: Juliana Fernandes Pavoni
• Host Institution: Faculdade de Filosofia, Ciências e Letras de Ribeirão Preto (FFCLRP). Universidade de São Paulo (USP). Ribeirão Preto , SP, Brazil
• Associated researchers: Gustavo Viani Arruda ; Oswaldo Baffa Filho

Abstract

Modern radiation therapy techniques deliver high doses of radiation to the target volumes of treatment, avoiding the healthy tissues involved. Many software are used in radiotherapy services since the initial consultation of the treatment, going through its planning, application, and finally arriving at the monitoring of the results. Failures along this process can result in serious errors, even compromising the lives of the patients involved. Thus, based on the extensive databases of these computerized systems, a perfect environment is created to apply artificial intelligence tools. In this context, this project aims to develop models for the prediction of approval results in the gamma analysis of the quality control tests of IMRT pre-treatment verification of patients treated with radiotherapy. We plan to use machine learning techniques and quantitative information about the treatment plan complexity, which can be automatically extracted from the treatment planning system. Besides, there is a consensus that the verification of dose distributions to be delivered to patients is necessary and that their measurement in three dimensions would be adequate. The main dosimeter for this purpose is the gel dosimeter, but its use is still restricted in clinical practice. To facilitate the gel dosimetry workflow and save time involved in the dosimeter calibration step, this project also aims to apply machine learning to automate the dosimeter calibration process. The dosimeter response curve would be predicted based on the quantitative characteristics of the non-irradiated gel sample's images. (AU)

Scientific publications

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

TOZZI, K.; LIZAR, J.; PAVONI, J.. Machine-Learning Prediction of the Gel Dosimetry Calibration Curve Equation Based On the Non-Irradiated Gel Magnetic Resonance Imaging. Medical Physics, v. 49, n. 6, p. 2-pg., 2022-06-01. (21/02254-6)

ROJAS, DIANA M. C.; PAVONI, JULIANA F.; ARRUDA, GUSTAVO V.; BAFFA, OSWALDO. Gel and thermoluminescence dosimetry for dose verifications of a real anatomy simulated prostate conformal radiation treatment in the presence of metallic femoral prosthesis. JOURNAL OF APPLIED CLINICAL MEDICAL PHYSICS, AUG 2021. (21/02254-6, 13/07699-0)

DA SILVEIRA, MATHEUS ANTONIO; PAVONI, JULIANA FERNANDES; BRUNO, ALEXANDRE COLELLO; ARRUDA, GUSTAVO VIANI; BAFFA, OSWALDO. Three-Dimensional Dosimetry by Optical-CT and Radiochromic Gel Dosimeter of a Multiple Isocenter Craniospinal Radiation Therapy Procedure. GELS, v. 8, n. 9, p. 12-pg., 2022-09-01. (21/02254-6, 13/07699-0)